Last updated: October 15, 2026
Every app, every server, every data centre and every connected device consumes resources. As organisations scale out their cloud environments, store more data and increasingly adopt AI powered technologies the issue of the environmental impact of IT becomes ever more pertinent.
Green computing is about making technology more resource efficient throughout its lifecycle; from the procurement of the hardware and the development of the software right through to the operation of the infrastructure and the eventual disposal of the equipment.
For businesses, the benefits can go beyond sustainability. Efficient IT systems may also reduce operating costs, improve infrastructure utilization, and simplify technology management.
Here are seven green computing best practices organizations may adopt to build a more eco-friendly IT environment.
Green computing is the practice of designing, using, managing, and disposing of computing in a way that minimizes its environmental impact.
It includes:
Green computing is not only for large data centers. Small businesses, software companies, and SMEs can also make their technology use more efficiently.
Digital infrastructure underpins pretty much every business function these days. Running that infrastructure requires energy, hardware and all the other supporting resources.
The IEA estimated global data centre electricity consumption at around 415 TWh in 2024. Its 2025 Energy and AI report projected that figure could be as high as 945 TWh by 2030 under its base case scenario.
Real demand will be a function of a number of factors including the level of AI adoption, technology improvements, infrastructure expansion and the availability of energy.
These trends make effective technology management more and more pertinent.
Organizations adopting green computing practices can work toward:
The best approach is to start with an understanding of the current resource usage rather than simply assuming that every new technology is going to be more environmentally friendly.
Choosing energy-efficient hardware is an important part of sustainable IT management.
Computers, servers, networking equipment, and peripherals differ in how efficiently they use electricity. When purchasing new devices, organizations should evaluate performance requirements alongside energy consumption.
For eligible products, ENERGY STAR certification can help identify equipment that meets applicable energy-efficiency criteria.
However, buying the most energy-efficient new device is not always the most sustainable decision.
Manufacturing the hardware itself has an energy cost, requires raw materials and involves transportation. The environmental cost of prematurely replacing equipment can in some cases negate some of the benefits of improved operating efficiency.
Practical steps:
Example: If an organization plans to replace its desktop computers, it can first check which devices still meet employee needs. Keeping equipment that still works well and only replacing inefficient or outdated devices is a balanced approach.
What to measure: device energy consumption, hardware utilization, replacement frequency, and equipment durability.
Equipment doesn’t need to be running at full power all the time.
Workstations, monitors, printers, and all the other assorted bits of kit can still be drawing power even when staff isn’t actively using them.
Configuring the right power management settings is one of the easiest ways to reduce avoidable electricity consumption.
Practical steps:
IT teams should be thinking about devices that require continuous availability, remote management, scheduled updates and security monitoring.
A blanket shutdown policy may not be appropriate in every environment.
Example: A business with hundreds of employee laptops could centrally manage their power settings rather than have each employee individually configure them.
What to measure: electricity consumption, active versus idle time and the % of eligible devices adhering to power management policies.
Electronic waste comprises discarded computers, servers, storage devices, mobile phones and all other electronic equipment.
Responsible equipment management should start before the equipment becomes waste.
Where possible organisations can extend the life of their hardware through maintenance, repair, upgrades, internal reassignment or refurbishment.
When equipment is no longer reusable appropriate recycling will recover the valuable materials and deal with the hazardous components.
Practical steps:
For organizations handling customer information or sensitive business records, secure data sanitization is essential before equipment is transferred, recycled, or disposed of.
Example: An organization replacing office laptops may be able to refurbish selected devices for less demanding internal tasks instead of purchasing new hardware for every employee.
What to measure: Equipment reuse rate, repair rate, devices recycled through approved channels, and average hardware lifespan.
Hardware is only one part of the environmental impact of IT.
Software architecture and design decisions influence how much computing capacity, memory, storage, network bandwidth, and supporting infrastructure an application requires.
Sustainable software development considers resource efficiency throughout the software lifecycle.
This includes architecture decisions, infrastructure requirements, performance testing, deployment, and ongoing maintenance.
Practical steps:
For example, an application that makes frequent, unnecessary API requests may consume more network and computing resources than required.
Improving the design can reduce those requests without diminishing functionality.
It is also important to avoid unnecessary complexity. Adding more services, automation, or infrastructure does not always produce a more efficient system.
What to measure: Resource consumption per transaction, application response times, infrastructure requirements, and performance under representative workloads.
Green coding focuses on writing and tuning software to use computing resources efficiently as it maintains the required functionality, reliability, and user experience.
Although closely related to sustainable software development, green coding concentrates more directly on implementation decisions.
Developers can have an impact on resource usage through their algorithms, their database queries, their memory management and the way their application behaves.
Practical steps:
Example: A reporting app was repeatedly scanning a large database table to produce the same summary. The database load and response time could have been reduced by optimising the query and introducing appropriate caching.
However faster code does not necessarily mean lower total energy consumption. The improvements need to be measured under the same conditions.
What to measure: CPU time, memory usage, database load, network transfer, execution time and energy consumption (where appropriate measurement tools are available).
Organizations continuously generate information through applications, transactions, logs, documents, backups, and connected systems.
Not all of this information remains useful.
Dark data generally refers to information an organization collects or stores but does not currently use for business, analytics, or other intended purposes.
Storing and retaining unnecessary information can increase storage requirements, backup volumes, management complexity, and associated costs.
Data should not be deleted simply because it’s infrequently accessed. There will be some data that needs to be retained for legal, compliance, audit, security or business continuity reasons.
Practical steps:
Example: A SaaS company may retain years of temporary exports and duplicate application logs. Reviewing these datasets could identify files that can be safely removed or moved to more suitable storage, subject to retention requirements.
What to measure: storage volume, storage growth, duplicated data, retained backups and storage cost.
Data centres support cloud apps, enterprise systems, digital services and AI workloads.
Their electricity consumption includes computing equipment and supporting infrastructure such as cooling, networking, and power systems.
Improving infrastructure efficiency requires more than purchasing newer servers.
Organisations need to know what resources they require, how effectively they are being used and where there is avoidable consumption.
Practical steps:
AI assisted monitoring and optimisation tools may be of use to the operator in identifying opportunities to improve cooling, capacity planning or resource allocation.
However they should be assessed in terms of actual results (including their own compute requirements).
Example: A cloud based application might have several very large VMs running all the time even though there’s very low demand overnight. Some usage monitoring and appropriate right-sizing would go a long way to reducing both unnecessary consumption and operating costs.
Organisations running their own data centres can also monitor their Power Usage Effectiveness (PUE) which is the ratio of total facility power to IT equipment power.
PUE is useful in terms of understanding facility overhead but on its own it doesn’t measure the environmental impact or efficiency of the application.
What to measure: cloud utilisation, electricity consumption, idle resources, infrastructure costs and relevant data centre efficiency metrics.
Knowing the best practices is only the starting point. You also need a practical way to prioritise the improvements.
A useful initial approach consists of five steps.
Step 1: Understand Current Resource Consumption
Check the organisation’s hardware inventory, cloud infrastructure, software workloads, storage utilisation and energy usage.
Find the areas where the resources are underemployed or where consumption is excessive.
Step 2: Prioritize High-Impact Opportunities
Not every sustainability initiative will have the same outcome.
Begin with opportunities that are technically feasible, quantifiable and in line with business requirements.
Examples would be the removal of idle cloud resources, the optimisation of inefficient database queries and the extension of the life of appropriate hardware.
Step 3: Establish Measurable Goals
Define clear objectives e.g. reduce idle infrastructure, improve server utilisation, reduce resource consumption per completed transaction.
Don’t use cost savings as proof of reduced environmental impact.
Step 4: Make Improvements Incrementally
Introduce the changes in small stages and check that they don’t have a negative impact on reliability, security, accessibility or the user experience.
Step 5: Monitor and Review Results
Compare resource usage pre and post each improvement.
Document the results and use the findings to inform future decisions.
An incremental, evidence-based approach helps organisations move beyond the ‘sustainability’ claims to actual, measurable improvements in their operations.
Different initiatives call for different metrics.
| Area | Useful Metrics |
| Hardware | Device energy consumption, utilization, useful life |
| Workplace IT | Electricity usage, idle time, power policy adoption |
| Software | CPU usage, memory consumption, energy or resources per transaction |
| Cloud infrastructure | Resource utilization, idle capacity, compute hours |
| Data management | Storage volume, backup volume, storage growth |
| Data centers | Electricity consumption, PUE, cooling demand |
| Environmental impact | Estimated greenhouse gas emissions using a documented calculation method |
The Green Software Foundation’s Software Carbon Intensity specification provides a methodology for evaluating software carbon emissions relative to a functional unit.
For organizations adopting formal sustainability measurements, consistent boundaries and assumptions are important for making meaningful comparisons.
Common green computing practices are: selecting energy efficient hardware, turning on power management, reusing and recycling electronics responsibly, writing efficient software, optimising code, managing data storage (i.e. not storing data you don’t need), and improving the way you use cloud and data centres.
Green computing covers the environmental impact of computing technologies across the entire spectrum of hardware, software, infrastructure and the equipment lifecycle.
Green coding is a more focused practice involving software implementation choices that aim to reduce unnecessary resource consumption.
Cloud platforms supply tools for resource monitoring, autoscaling, and infrastructure optimization. These capabilities can help organizations reduce unused computing capacity.
However, cloud migration does not automatically diminish environmental impact. Results depend on architecture, resource utilization, infrastructure efficiency, and electricity sources.
Not necessarily. The new hardware may in fact use less power when in operation, but the manufacture and replacement of the hardware will have its own environmental impact. You need to look at the whole lifecycle of the device and how it’s actually being used before you make the decision to replace the existing equipment.
Dark data is data which has been collected or stored by an organisation and is not currently being used for its intended purpose.
Managing the data you don’t need can go a long way to reducing your storage requirements. Of course your retention decisions will have to take into account business, legal and security considerations.
AI based tools could assist with cooling optimisation, resource forecasting and operational monitoring. The benefits would be dependent upon the infrastructure, the way they are implemented and the actual savings that are achieved.
No. Businesses of all sizes can improve power management, extend the life of their devices, optimize their applications and manage their cloud and storage resources more effectively.
Green computing is not a single technology or one-time initiative. It involves making better decisions about how software, hardware, data, and infrastructure are designed, used, and maintained.
The best improvements typically start with the systems you already have in place: finding out what’s being wasted, where you can extend the life of existing equipment, how to make your software more efficient and how to get the most out of your cloud workloads.
By setting measurable goals and evaluating actual outcomes, organizations can improve resource efficiency while supporting business performance.
Build More Efficient Software and IT Systems with InApp
Improving technology efficiency often requires changes to software architecture, legacy applications, cloud infrastructure, and development practices.
InApp helps organizations modernize applications, optimize cloud environments, and develop software systems suited to developing business requirements.
